APPLICANT'S ABSTRACT: The proposed research will utilize standard behavioral-pharmacologic methods of double blind drug/placebo preference for the study of alprazolam self-medication behavior in anxious outpatients who, because of their psychopathology, are presumed to be at risk for benzodiazepine dependence. All patients will have generalized anxiety but will have no history of substance abuse or dependence. Patients will be allowed to use medication "as needed" and self-administration behaviors will be monitored through the use of automated recording devices. Following initial exposures to color-coded placebo and alprazolam capsules, subjects will be able to choose, under double-blind conditions, which capsules they prefer. The basic hypothesis of this proposal is that the anxiolytic efficacy of alprazolam will reinforce self-medication behavior in anxious patients but that medication use generally will be therapeutically appropriate and will correspond with anxiety symptom relief. Two experiments will employ patients who have little or no history of prescription anxiolytic use. These studies will: 1) demonstrate reinforcing effects of alprazolam in generally anxious patients with or without panic disorder; 2) determine the effect of patient drug use history on prescription drug use; and 3) determine the ability of cognitive behavior therapy to reduce anxiety symptoms and minimize drug use in anxiety disorders. Both experiments will assess the ability of personality type and patient attitude towards medication use to predict observed drug self-administration behaviors. This research is consistent with NIDA's goals to examine the basic mechanisms of drug reinforcement and the risk factors making one vulnerable to prescription drug dependence. There are three factors of vulnerability being experimentally examined. These are: 1) patient history of other drug use; 2) personality type and attitudes towards drug use; and 3) the effects of untreated anxiety psychopathology to promote self-medication. Furthermore, a significant accomplishment of this research will be the development of outpatient procedures for the study of prescription drug self-administration and reinforcement in therapeutic populations. The results will address fundamental questions regarding the self-medication motivation and its role in the etiology of prescription drug dependence. Furthermore, these studies may experimentally identify patient variables which can predict patterns of medication use placing patients at risk for benzodiazepine dependence. Finally, the proposed research will determine the ability of cognitive behavior therapy to reduce self-medication drug use behaviors and thereby minimize the risks of benzodiazepine dependence.